35
votes

NOTICE:

Since this site is now in beta, it is time to have these conversations with the folks who are actually building this site. Please bring your comments and discussions to this site's 'meta' support forum so not to split these conversations. Thanks.

Proposal: Emacs

Emacs is on topic at SuperUser, with over 1350 questions asked and answered. Why would we need or support a separate site? Are there any questions proposed that would not be on topic at Superuser?

1
  • The question title is much broader than the body. +1 to the title (a separate site will not help anything, AFAICT). -1 to the body: it's not about SuperUser - see @Warren's answer. Please consider editing the body to make it as general as the title.
    – Drew
    Sep 20, 2014 at 15:27

5 Answers 5

Reset to default
38
votes

Because emacs questions aren't only on superuser, they are on programmers.stackexchange, stackoverflow, tex and probably more. Likewise, emacs knowledge, solutions are equally scattered on said sites plus on #emacs, github, youtube.

In short, they are scattered all over the place.

Also, superusers questions might not include specific elisp issues.

7
  • 12
    "Scattered all over the place" because Emacs is a tool, and you ask your question based on the expertise you are trying to apply. That's not a reason "for" or "against" exploring this proposal, but I'm not sure "scattered all over the place" is the argument I would use for creating yet another site.
    – Robert Cartaino
    Sep 10, 2014 at 18:19
  • 5
    emacs is also a programming platform. Programming questions were also scattered all over the place before stackoverflow, certainly that was a factor in creating a centralized place like stackoverflow.
    – c-o-d
    Sep 10, 2014 at 18:41
  • 2
    @EdgarAroutiounian - based on your comment, it's safe to conclude you think emacs questions belong on Stack overflow :)
    – warren
    Sep 11, 2014 at 2:57
  • 3
    @warren I don't think its safe to conclude that at all. Just as ubuntu specific questions are better handled on askubuntu, so too would emacs related questions be better handled on a dedicated emacs stackexchange. Another example: tex is turing complete, but we ask tex related questions on tex.stackexchange, not stackoverflow.
    – c-o-d
    Sep 11, 2014 at 3:00
  • 3
    This seems like a reason against creating yet-another-site...
    – BlueRaja - Danny Pflughoeft
    Sep 16, 2014 at 0:39
  • 1
    "Scattered all over the place" is a slipper slope argument. Samba, Bash, AWK, Perl, are scatter all over those same sites. Shall we make everything have its own SE site? I think people need to use the tools that already exist better, or be made more aware of them. You can search across the various SE sites using the search engine on stackexchange.com or Google, if you're looking for a common thing across the SE sites.
    – slm
    Sep 16, 2014 at 11:39
  • Filter on tag [emacs] or [elisp] or both. End of story. You won't miss anything. The "scattering" fear is a bogeyman.
    – Drew
    Sep 20, 2014 at 15:24
23
votes

My first thought when I heard about this proposal was

xkcd standards

We now have stackoverflow for emacs+dev, unix for emacs+unix, superuser for emacs+user, so why add a fourth side (or maybe fifth, sixth if you count the other sides where emacs questions could be on-topic)? Then again, I remembered how the situation was with tex, which was also borderline stackoverflow/superuser, and now we have tex.stackexchange.com which is a huge success.

Before tex.stackexchange the tex questions where scattered all over the stackexchange network. Now there is a single, highly successful site for those questions. I think the chances are good that the same will happen with emacs.stackexchange.com. That's why we need emacs.stackexchange.

5
  • 5
    The parallel with tex is a very apt one.
    – Erik P.
    Sep 11, 2014 at 14:03
  • I hope you're correct.
    – slm
    Sep 16, 2014 at 11:28
  • The TeX community really seems to have taken to SE. They don't seem concerned by the possibility that the site is proprietary and could just disappear at any time. Part of this reason is that TeX, despite being written in text, is in part inherently visual/pictorial, therefore is ideally suited to a non-entirely-text medium. In particular, with the rise of PGF/TikZ, more people are using TeX to draw vector graphics.
    – Faheem Mitha
    Sep 16, 2014 at 13:13
  • 2
    "...that the site is proprietary and could just disappear at any time." That is not entirely true, while the site itself is proprietary, the data is not. IIRC you can get a full data dump of any stackexchange site.
    – Flow
    Sep 16, 2014 at 14:33
  • 1
    @Flow And, more importantly, you can actually use the full data dump to seed a different site that is fully open source and non-profit.
    – gerrit
    Sep 22, 2014 at 18:06
14
votes

Super User is not the best place to ask Emacs questions—Stack Overflow is. Over on the little-used Stack Exchange homepage, you can see a filter of all questions tagged "emacs". It shows the network gets roughly 7 Emacs questions a day. If you add in elisp and other emacs-specific tags, the results are nearly identical. The vast majority of questions are hosted on Stack Overflow.

The test case for splitting off a subset of a larger site is the Unix-Ubuntu split:

I don't know if there is a general rule to be derived from this case study. I’m still a big fan of the larger sites. I'm wary of encouraging smaller, focused sites when a larger group stands to benefit from their combined resources. Stackoverflow.com is a testament to how well diverse communities can pool their resources when they have a common goal; in this case, writing better code. Maybe that’s a criteria which binds a community together: do they have common goals? Stack Overflow didn't split off into a .NET site, a PHP site, a C# site, a Java site, etc… and we're better off for the experience.

I've used Emacs (actually XEmacs) to do all sorts of things:

  • read email and newsgroups,
  • IRC,
  • terminal window,
  • web browser,
  • general editor,
  • gaming,
  • playing music (though to be honest, this never worked for me),
  • Lisp sandbox,
  • development environment, and etc.

For several years, I practically lived in Emacs frames. So in theory Emacs questions could be on topic on dozens of sites. But in practice, most Emacs questions are, deep down, Elisp questions. And that means their goals are aligned with Stack Overflow's goal: "writing better code". A quick scan of Emacs questions (whether on SO or elsewhere) show that nearly all of them are about how to make it work better as an IDE.

2
  • 3
    Problem being that much of the time we don't care much about the code. We're merely trying to use code that others have written. The only 'code' we're writing are the calls necessary to plug it in. The hard part is plugging it in to the rest of the ecosystem (ex: 'how do I set up multiple accounts in mu4e?'; answer: some emacs config, other program config, and elisp. 2 of the 3 don't fit on StackOverflow)
    – J David Smith
    Sep 10, 2014 at 21:56
  • 2
    I agree with David, most questions about Emacs is how to get a task done with Emacs. For example, blog.binchen.org/posts/… , I discussed a little bit git, bash code, firefox plugin, emacs plugin. It's hard to categorize the post into stackoverflow or serverfault
    – chen bin
    Sep 11, 2014 at 5:29
10
votes

We don't - and this has been proposed several times in the past (just like proposals for vi[m]) - they're all gone now, but they've appeared time and time again over the life of Area51.

Where emacs is already covered:

10
  • 15
    I think listing six different places that have significant emacs questions shows it's a topic looking for a home.
    – aaronosaurus
    Sep 10, 2014 at 17:23
  • 4
    @Anonymous - only one truly has significant population - and that's SO. SU is second, the others are minor, and probably should be moved to SU or SO.
    – warren
    Sep 10, 2014 at 17:27
  • 3
    On 'Programmers' around half the questions are closed, and on ServerFault almost all would be better elsewhere.
    – DJClayworth
    Sep 10, 2014 at 18:02
  • 1
    Precisely! You have listed six places people need to think to search or post if they have a question along the lines of "how do I do X in emacs", which results in duplicates, closed questions and worst of all—unsolved user problems.
    – Mark
    Sep 16, 2014 at 1:32
  • 2
    @MarkAufflick - there's really only one place: Stack Overflow
    – warren
    Sep 16, 2014 at 1:42
  • 1
    @warren then why are there > 1300 questions on SU? There are also 329 results on SO for "emacs closed:yes duplicate:no" (admittedly some for unrelated normal *O reasons).
    – Mark
    Sep 16, 2014 at 2:53
  • 2
    @aaronosaurus - I think you missed the point of the links above. Look at the counts. We're creating another SE site to contain ~13k Q's which, oh by the way, 11k are already nicely contained on SO.
    – slm
    Sep 16, 2014 at 11:31
  • 1
    1. Yes to @slm's comment. Please put the SO count in bold or something, so it is more obvious. 84% are on SO. 2. It is trivial to filter on tags [emacs] and [elisp] across all sites, if you want to find Emacs-related questions.
    – Drew
    Sep 20, 2014 at 15:20
  • 1
    It is trivial to filter if you are looking for an answer to an already-asked question. If you want to ask a question, on the other hand, you still have to pick one site.
    – T. Verron
    Sep 22, 2014 at 6:02
  • @T.Verron And often they're on the border between application questions and programming questions, which sometimes gets them closed - though they might still be one of the top hits on Google. A link to other relevant exchange sites under the 'This question is closed' message would be helpful though.
    – Brian Burns
    Sep 23, 2014 at 4:48
0
votes

To me, it is not about "can Stackoverflow already fill the need for an emacs sx?", but rather "does it?". And this question is not about "where can we find answers about emacs?", it is about "where can we ask questions about emacs?".

The emacs tag on stackoverflow is running at ~10 questions per day. It would be within the "okay" range for a beta website, but given the surge of interest for this very beta, I wouldn't be surprised if we ended with way more than these ~10 questions.

Note also that it is very rarely that emacs questions to google lead back to stackoverflow (or any other stackexchange site). Instead you have blogs, you have emacswiki, you have scattered wikis on github, and you have gmane (which is underrepresented on google as well, though).

So this leads me to believe that there is a need for an emacs Q/A website. Stackoverflow, superuser, etc. may be technically able to fulfill that need, but apparently they don't (otherwise, there wouldn't be a need anymore, eh?).

So the question is "why aren't there more emacs-related questions on SO?" I do not have a definite answer to that, but a few points were already mentionned around the place:

  • people may prefer asking questions in a more community-like environment (hell, there are questions on emacswiki ! ), and you can't really build a community around a tag;
  • SO is really intimidating. The amount of questions, the pace at which they appear and disappear, the very technical appearance of the questions on the first page... It makes it very hard to dive in for newcomers, especially with questions which wouldn't look quite right (how many emacs questions do you see on the front page? ~10 questions a day is nothing for SO!);
  • not all emacs users are professional programmers. Some are amateur programmers, some are researchers in various fields... The same way TeX is trying to appeal to more than just programmers and mathematicians, a lot of people could benefit from using emacs. Students taking notes with MS Word could use org-mode, no matter if it is a math or a history class! Do you really see a history student asking a question on SO? :p

The second of these problems could be solved with appropriate information and "marketing", of the form "questions like XXX are also welcome on SO". But then a lot of other underrepresented communities would rightfully be annoyed that they don't get this kind of publicity as well (but most of these communities could not populate an independent SX site of their own). And the other problems would remain unsolved.

Several people have already made a parallel with tex.sx vs SO and askubuntu vs UX. This parallel extends to the three problems I have stated above. And as far as I know, both these sites ended up nicely complementing the set of advice that the older, bigger websites could provide, and, most importantly, they have managed to fill the niche left by the third problem: beginners are welcome and feel welcome.

Tl;dr: An emacs.sx website will not divert much traffic/questions/answers from SO and other sites, because this traffic is already diverted by a dozen blogs, by emacswiki and by gmane. Yet apparently a lot of SX-ers would be willing to use an emacs.sx website.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .